PPE Image Annotation Specialist (Object Detection & Polygon)

About OpenTrain

OpenTrain is the #1 platform for finding and building careers in AI training and data labeling. We connect skilled annotators with practical AI projects so contributors can learn, earn, and shape how modern models behave.

About AI Training Work

AI training (data labeling/annotation) is the human side of building intelligent systems: people create the examples models learn from. This work is flexible, usually fully remote, and accessible to contributors with attention to detail or domain knowledge.

On this project you'll be directly improving a computer vision system that detects Personal Protective Equipment (PPE) in images—work that helps make AI safer and more useful in real-world safety contexts.

Project Overview

We need an experienced image annotation specialist to label images for a PPE detection model. This is a fixed-price contractor role with a total payment of USD 2,000. Work is done remotely and is open worldwide.

Label types required: object detection (bounding boxes) and polygon masks. The designated annotation tool for this project is Roboflow; previous experience with CVAT, Label Studio, Roboflow, or similar tools is expected.

Annotation Classes

Annotate each object in the image with accurate bounding boxes or polygon outlines and assign one of the classes below.

• Gloves
• Hardhat
• NO-Gloves
• NO-Hardhat
• No-Vest-Overall
• Overall
• Person
• Safety Vest

What You'll Do

Perform high-quality image annotation across the provided dataset using Roboflow and the specified label types. Maintain consistency and precision so annotations can be used directly for model training and evaluation.

• Create accurate bounding boxes and polygon masks for each object instance.
• Assign the correct class label from the predefined set for every annotated object.
• Follow annotation guidelines and quality checks to ensure consistency across images.
• Use Roboflow for annotation workflows and deliver labeled datasets in the required format.

Requirements

You must be an expert labeler with proven experience and a track record of delivering high-accuracy object-detection datasets.

• Proven experience with image annotation tools (e.g., CVAT, Label Studio, Roboflow).
• Demonstrated ability to maintain high accuracy and consistency across images.
• Experience with object detection and polygon labeling workflows.
• Familiarity with AI datasets for object detection is a plus.

How It Works & Compensation

This is a contract role with a fixed total payment of USD 2,000. OpenTrain AI manages contracting, quality review, and payment. The role is remote and open worldwide; hours and schedule are set between the contractor and project manager.

Apply through OpenTrain, share samples or portfolio of prior object-detection work if available, and you will receive task instructions and annotation guidelines after selection.

• Employment type: Contractor (fixed-price contract).
• Labeling software: Roboflow; label types: OBJECT_DETECTION and POLYGON.
• Total payment: USD 2,000 (fixed price).
• Worldwide applicants welcome; language and location requirements are not specified.